Farol: Static Risk Analysis for Pull Requests Prior to Merge

Repositorywinnin/farolTool

Farol is an open-source CLI and GitHub Action designed to evaluate the risk of pull requests. Using static analysis, it maps call chains, diff test coverage, and contract breaks to help human reviewers focus their attention on the most dangerous changes.

Key points

  1. Complements AI code reviews using deterministic compiler-based static analysis instead of LLM estimates.
  2. Assigns a 0–100 risk score and tier verdict: fast track, quick review, or careful review.
  3. Maps blast radius across affected pages, API endpoints, jobs, and RPC methods with complete call chains.
  4. Supports TypeScript/React (Next.js), Go, and Python (FastAPI).
  5. Runs on Node.js 22+ via npx or as a GitHub Action with non-blocking shadow mode support.

Why it matters

As AI agents generate more code, human approval becomes the primary bottleneck; Farol ensures review effort scales with the architectural risk of a change rather than diff size alone.

Analysis

The Problem: Review Fatigue in Agentic Workflows

When coding agents write diffs and automated agents review them, human approvers often skim PRs evenly. This superficial triage risks letting small modifications to shared components break downstream dependencies across dozens of routes.

How Farol Works

Farol does not replace automated reviews; it operates downstream of them using compiler-derived static analysis:

  • Blast Radius Mapping: Determines every page, API route, or background task executing the changed logic.
  • Diff Coverage & Contract Checks: Identifies uncovered lines inside the diff and flags broken API contracts.
  • Review Prioritisation: Generates an ordered guide showing human reviewers where to inspect code first.

Maturity and Target Audience

Farol is a relatively young project (155 stars, MIT licence) targeting engineering teams maintaining web applications and microservices in TypeScript, Go, or Python. Its shadow mode enables non-disruptive CI evaluation before enforcing risk gates.

Mentioned links

Found in

Summary generated by AI (Gemini 3.8 Flash, 28 October 2026). It may contain errors – the original source is authoritative.

Related entries

Similar in content, found by the AI search.

  • Repository:coldteadotai/pr-lens

    PR Lens: Animated Architecture and Data-Flow Diagrams for Pull Requests

    PR Lens visualises changes in GitHub pull requests as animated architecture and data-flow diagrams posted directly as comments. It enables developers to grasp system-level impacts before conducting line-by-line code reviews.

    960starsTypeScript

    Web development· Tool

  • Repository:prathamdby/pr-agent

    PR Agent: Self-Hosted AI Reviewer for Pull Requests

    PR Agent is an open-source, TypeScript-based automated code review tool designed to run on self-owned infrastructure as a self-hosted alternative to per-seat SaaS services like CodeRabbit.

    46starsTypeScript

    AI & LLMs· Tool

  • Article:The Cognition Team

    Devin Review: Cognition Unveils AI-Assisted Code Review Tool

    Cognition has introduced Devin Review, an AI-powered code review tool designed to help developers inspect and understand increasingly complex pull requests produced by human engineers and coding agents.

    AI & LLMs· Announcement

  • Repository:earendil-works/pi-review

    pi-review: Code Review Extension for Pi

    pi-review is an open-source extension for the Pi tool that integrates structured code reviews directly into the command-line interface. It analyses uncommitted changes, branches, commits, GitHub pull requests, or directory snapshots to generate prioritised findings and actionable follow-ups.

    549starsTypeScript

    AI & LLMs· Tool

  • Link:davis7.link

    Greptile: AI Code Review Platform Using Graph Indexing and Agent Swarms

    Greptile provides automated pull request reviews by indexing codebases into a dependency graph and deploying a swarm of parallel agents. It aims to catch logic bugs, security issues, and style violations across entire repositories rather than just isolated diffs.

    AI & LLMs· Tool

Back to the knowledge base

Let's talk about your project

Get in touch

Locations

  • Mattersburg
    Johann Nepomuk Bergerstraße 7/2/14
    7210 Mattersburg, Austria
  • Vienna
    Ungargasse 64-66/3/404
    1030 Wien, Austria

Parts of this content were created with the assistance of AI.